Skills
98,067 packaged capabilities and playbooks for agents. Install any of them with one command.
Development
Use when writing or reviewing JavaScript/TypeScript in this repo that builds an interactive voice agent via `agent.deepgram.com/v1/agent/converse`. Covers `client.agent.v1.createConnection()` / `connect()`, `sendSettings`, `sendMedia`, runtime updates, event handling, and function-call responses. Use `deepgram-js-text-to-speech` for one-way synthesis, `deepgram-js-speech-to-text` or `deepgram-js-conversational-stt` for transcription only, and `deepgram-js-management-api` for project/model admin rather than live agent runtime. Triggers include "voice agent", "agent converse", "full duplex", "barge-in", "function calling", and "agent.v1".
Interview the user to define or revise the app's overall product direction, then preserve the confirmed meaning in root PRODUCT.md. Always use for a new app direction or any app-wide change to users, usage situations, problem, promised change, core loop, durable boundaries, experience principles, success signals, assumptions, or unknowns, even when PRODUCT.md is not named. Start from an existing definition or a rough app direction. Do not use for blank-page idea discovery, architecture, or shaping or specifying one feature or work unit.
Use when working with Fresh framework, creating routes or handlers in Fresh, building web UIs with Preact, or adding Tailwind CSS in Deno. Covers Fresh 2.x project structure, route handlers, islands, createDefine, PageProps, context patterns, and Fresh 1.x to 2.x migration. Essential for any Fresh-related question.
Apply Gang of Four design patterns to solve architectural problems in TypeScript. Use when refactoring code architecture, implementing extensible systems, decoupling components, or following SOLID principles. Covers all 22 GoF patterns: Creational (Factory Method, Abstract Factory, Builder, Prototype, Singleton), Structural (Adapter, Bridge, Composite, Decorator, Facade, Flyweight, Proxy), and Behavioral (Chain of Responsibility, Command, Iterator, Mediator, Memento, Observer, State, Strategy, Template Method, Visitor).
Recover the original logic from code protected by a virtualization obfuscator — VMProtect, Themida/WinLicense, Code Virtualizer, or a custom opcode VM — by locating the VM dispatcher, reverse-engineering the handlers into semantics, extracting the virtual bytecode, and lifting it to a simplified IR with Triton, miasm, or VTIL-based tools. Use when a function became a giant fetch-decode-dispatch loop, when analysis shows a handler table instead of normal code, or after unpacking reveals a virtualized core.
Provides PyTorch-native distributed LLM pretraining using torchtitan with 4D parallelism (FSDP2, TP, PP, CP). Use when pretraining Llama 3.1, DeepSeek V3, or custom models at scale from 8 to 512+ GPUs with Float8, torch.compile, and distributed checkpointing.
Drupal contrib module upgrade triage — Drupal.org issue queue research, local patch creation (git diff), composer-patches workflows, Drupal 11 compatibility via core_version_requirement, Drupal Lenient handling, and upgrade_status analysis. Not for generic Composer dependency management.
Builds and optimizes language model programs with DSPy (Stanford NLP), using Signatures, modules (Predict, ChainOfThought, ReAct, ProgramOfThought) and optimizers (BootstrapFewShot, MIPRO, BootstrapFinetune). Use when replacing hand-written prompts with declarative signatures, when automatically tuning prompts against training data and a metric, when building multi-stage RAG pipelines or ReAct agents, when building classifiers or structured-output programs, or when saving and evaluating optimized LM programs across providers such as Anthropic, OpenAI, or Ollama. Not for quick prototypes or simple chains; use manual prompting or LangChain instead.
Build conversational AI voice agents with ElevenLabs Platform using React, JavaScript, React Native, or Swift SDKs. Configure agents, tools (client/server/MCP), RAG knowledge bases, multi-voice, and Scribe real-time STT. Use when: building voice chat interfaces, implementing AI phone agents with Twilio, configuring agent workflows or tools, adding RAG knowledge bases, testing with CLI "agents as code", or troubleshooting deprecated @11labs packages, Android audio cutoff, CSP violations, dynamic variables, or WebRTC config. Keywords: ElevenLabs Agents, ElevenLabs voice agents, AI voice agents, conversational AI, @elevenlabs/react, @elevenlabs/client, @elevenlabs/react-native, @elevenlabs/elevenlabs-js, @elevenlabs/agents-cli, elevenlabs SDK, voice AI, TTS, text-to-speech, ASR, speech recognition, turn-taking model, WebRTC voice, WebSocket voice, ElevenLabs conversation, agent system prompt, agent tools, agent knowledge base, RAG voice agents, multi-voice agents, pronunciation dictionary, voice speed control, elevenlabs scribe, @11labs deprecated, Android audio cutoff, CSP violation elevenlabs, dynamic variables elevenlabs, case-sensitive tool names, webhook authentication
Tracks ML experiments with SwanLab, an open-source tool covering swanlab.init, config and metric logging, scalar charts, and media logging (images, audio, text, GIFs, point clouds, molecules), in cloud, local (mode="local" with swanlab[dashboard]), or self-hosted setups. Includes integrations for PyTorch, Transformers, PyTorch Lightning, and Fastai. Use when logging metrics and hyperparameters for training runs, comparing runs across seeds, checkpoints, or hyperparameters, running an offline or self-hosted dashboard instead of managed SaaS, logging media alongside scalars, or adding tracking callbacks to a Transformers, Lightning, or Fastai training loop. Not for other trackers such as Weights & Biases or MLflow.
Fine-tune and serve Physical Intelligence OpenPI models (pi0, pi0-fast, pi0.5) using JAX or PyTorch backends for robot policy inference across ALOHA, DROID, and LIBERO environments. Use when adapting pi0 models to custom datasets, converting JAX checkpoints to PyTorch, running policy inference servers, or debugging norm stats and GPU memory issues
Fine-tune LLMs using reinforcement learning with TRL - SFT for instruction tuning, DPO for preference alignment, PPO/GRPO for reward optimization, and reward model training. Use when need RLHF, align model with preferences, or train from human feedback. Works with HuggingFace Transformers.
Design and implement purposeful, performant, and accessible motion in Flutter. Use for transitions, micro-interactions, gesture feedback, animated state changes, shared elements, list choreography, reduced-motion behavior, or diagnosing animation quality and performance.
Provides one-command project creation for Flutter including project initialization, configuration, and template generation. Use when the user asks about creating Flutter projects, needs to initialize a new Flutter project, or generate Flutter project structure.
Configure and use the SAP AI Core Orchestration service: build prompt templates, add content filtering, set up document grounding (RAG) pipelines, and call models via the Harmonized API. WHEN: user asks about the orchestration service, Harmonized API, prompt templating with variables, content filtering, grounding, RAG pipeline, document grounding, or building a pipeline that wraps an LLM. DO NOT USE FOR: creating an orchestration deployment — use `aicore-lifecycle-management` for that; credential setup — use `aicore-admin-resources`; browsing or deploying foundation models directly without a pipeline — use `genai-hub-foundation-models`; direct LLM inference without a pipeline — use `aicore-sdk-java` or `aicore-sdk-js`
One-time GitHub repository configuration via the `gh` CLI: enable a merge queue on the default branch, lock merging to squash-only with PR-title commit messages, wire required CI status checks, scaffold automatically-generated release notes (`.github/release.yml`), and optionally add a tag-driven release workflow. Use when the user says "set up the repo", "configure the merge queue", "enable squash merging", "lock down main", "set up branch protection", "configure release notes", "scaffold releases", "set up release tagging", or invokes /gh-bootstrap on a fresh or loosely-configured GitHub repository. Idempotent — safe to re-run on already-configured repos to bring missing pieces in line. Distinct from /gh, which handles per-task PR / issue / CI ops; this skill is for one-time repo policy.
Answer a numbered GitHub Discussion thread, read-only: read the thread and the repository to form an answer, then either post a single top-level reply or stop without commenting. Never executes code, never commits, never runs mutative git or gh commands, and never reads outside the repository. Use when the user invokes `/gh-discuss NUMBER`, `$gh-discuss NUMBER`, or asks to answer a GitHub Discussions thread.
Generates comprehensive GitHub activity reports across all authenticated accounts. Gathers repos, PRs, features, and themes for configurable time periods (1/5/7/30/90 days). Produces shareable markdown with tables, mermaid charts, and executive summaries. Can create a private repo with GitHub Actions automation and GitHub Pages aggregation site. Use when: "github report", "work report", "activity summary", "what did I work on", "gh-work-report", "show my github activity".
Generate conventional commit messages automatically. Use when user runs git commit, stages changes, or asks for commit message help. Analyzes git diff to create clear, descriptive conventional commit messages. Triggers on git commit, staged changes, commit message requests.
Git create MR. Use this skill when the user asks to create an MR, merge request, merge into a certain environment/branch, or any operation involving MR creation. Also applies to colloquial phrasings such as "make an MR", "submit a merge request", "merge to test", "merge to develop", "merge to production", "merge to the test environment", "merge to the development environment", "merge to the staging environment".
Manage and generate .gitignore files using the official GitHub gitignore template library via the `ign` CLI. Use when initializing new project repositories, automatically detecting project stack from files, adding language/framework/IDE ignore rules, inspecting applied templates, or safely modifying existing .gitignore files.
Build, scaffold, debug, and publish apps for the Homey smart home platform using the Homey Apps SDK v3. Use this skill whenever the user mentions Homey, Homey Pro, Homey Cloud, Homey app development, Homey CLI, Homey Compose, Homey Flow cards, Homey device drivers, Homey capabilities, Homey widgets, Homey pairing, or any smart home app development targeting the Homey platform. Also trigger when the user wants to integrate a device brand or cloud service with Homey, create Flow automations, build dashboard widgets, or publish to the Homey App Store. Even if the user just says "I want to make a Homey app" or mentions a device they want to control with Homey, use this skill.
Build type-safe APIs with Hono - fast, lightweight routing for Cloudflare Workers, Deno, Bun, and Node.js. Set up routing patterns, middleware composition, request validation (Zod/Valibot/Typia/ArkType), RPC client/server with full type inference, and error handling with HTTPException. Use when: building APIs with Hono, setting up request validation with schema libraries, creating type-safe RPC client/server communication, implementing custom middleware chains, handling errors with HTTPException, extending context with custom variables, or troubleshooting middleware type inference issues, validation hook confusion, RPC performance problems, or middleware response typing errors.
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.